Ethanol abuse and alcoholism remain very serious societal problems. A significant problem is the inability to diagnose alcohol abuse either in the general population or within selected groups of individuals such as adolescents and the recovering alcoholic. Accordingly, this proposal seeks to develop diagnostic biomarker signatures of acute and chronic alcohol consumption for diagnosing high-risk drinking, detecting relapse to drinking, disclosing recent drinking and in high risk situations such as pregnancy. To this end, studies are proposed to examine serum proteins and protein patterns for potential signatures in a powerful non-human primate model that is not encumbered by problems of comorbid drug use, inadequate diet and unreliable assessments of drinking history. In these NIAAA-funded, ongoing, within-subject studies, monkeys have been induced to voluntarily drink large amounts of alcohol. In the course of the studies (encompassing over 100 individual animals covering years of behavior and observation), serum samples have routinely been collected and archived. Experiments are proposed to screen these samples for potential biomarkers that can then be taken forward into the human population. Serum samples from a long-standing nonhuman primate self-administration study will be used as a training set for biomarker identification using high throughput proteomics. Samples will be processedto deplete the most abundant, obscuring proteins and then subjected to 2-DIGE (2-D Fluorescence Difference In-Gel Electrophoresis) for quantitative fluorescence identification of altered serum protein expression followed by MALDI-ToF/ToF identification of protein species. Statistical validation will be conducted, in a blinded fashion, using a test set of samples from an independent colony of self-administering monkeys, which will also contain data on adolescent vulnerability. The key criteria of any putative biomarkers will be sensitivity (percentage of positive scores among drinkers) and specificity (percentage of false positives in a non-drinking population). In addition, these studies will provide initial indices of positive and negative predictive values for biomarker signatures. A clinical test for ethanol abuse and alcoholism would have many potential uses. To discover protein biomarkers of ethanol abuse and alcoholism, serum from a controlled non-human primate population self- administering ethanol will be examined by quantitative proteomic methods.